corner-harris

corner_harris

skimage.feature.corner_harris(image, method='k', k=0.05, eps=1e-06, sigma=1) [source]

Compute Harris corner measure response image.

This corner detector uses information from the auto-correlation matrix A:

1 2	A = [(imx**2)   (imx*imy)] = [Axx Axy]     [(imx*imy)   (imy**2)]   [Axy Ayy]

Where imx and imy are first derivatives, averaged with a gaussian filter. The corner measure is then defined as:

1	det(A) - k * trace(A)**2

or:

1	2 * det(A) / (trace(A) + eps)

Parameters:

image : ndarray Input image. method : {‘k’, ‘eps’}, optional Method to compute the response image from the auto-correlation matrix. k : float, optional Sensitivity factor to separate corners from edges, typically in range [0, 0.2]. Small values of k result in detection of sharp corners. eps : float, optional Normalisation factor (Noble’s corner measure). sigma : float, optional Standard deviation used for the Gaussian kernel, which is used as weighting function for the auto-correlation matrix.

Returns:

response : ndarray Harris response image.

References

[R134]

http://kiwi.cs.dal.ca/~dparks/CornerDetection/harris.htm

[R135]

http://en.wikipedia.org/wiki/Corner_detection

Examples

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19

>>> from skimage.feature import corner_harris, corner_peaks >>> square = np.zeros([10, 10]) >>> square[2:8, 2:8] = 1 >>> square.astype(int) array([[0, 0, 0, 0, 0, 0, 0, 0, 0, 0],        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],        [0, 0, 1, 1, 1, 1, 1, 1, 0, 0],        [0, 0, 1, 1, 1, 1, 1, 1, 0, 0],        [0, 0, 1, 1, 1, 1, 1, 1, 0, 0],        [0, 0, 1, 1, 1, 1, 1, 1, 0, 0],        [0, 0, 1, 1, 1, 1, 1, 1, 0, 0],        [0, 0, 1, 1, 1, 1, 1, 1, 0, 0],        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0],        [0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]) >>> corner_peaks(corner_harris(square), min_distance=1) array([[2, 2],        [2, 7],        [7, 2],        [7, 7]])